Type:	Package
Title:	Interface to 'MLflow'
Version:	2.22.1
Maintainer:	Matei Zaharia <matei@databricks.com>
Description:	R interface to 'MLflow', open source platform for the complete machine learning life cycle, see https://mlflow.org/. This package supports installing 'MLflow', tracking experiments, creating and running projects, and saving and serving models.
License:	Apache License 2.0
URL:	https://github.com/mlflow/mlflow
BugReports:	https://github.com/mlflow/mlflow/issues
Depends:	R (≥ 3.3.0)
Imports:	base64enc, forge, fs, git2r, glue, httpuv, httr, ini, jsonlite, openssl, processx, purrr, rlang (≥ 0.2.0), swagger, tibble (≥ 2.0.0), withr, yaml, zeallot
Suggests:	carrier, covr, h2o, keras, lintr, sparklyr, stringi, testthat (≥ 2.0.0), reticulate, xgboost
Encoding:	UTF-8
RoxygenNote:	7.1.2
Collate:	'cli.R' 'databricks-utils.R' 'globals.R' 'imports.R' 'logging.R' 'mlflow-package.R' 'model-crate.R' 'model-python.R' 'model.R' 'model-utils.R' 'model-h2o.R' 'model-keras.R' 'model-registry.R' 'model-serve.R' 'model-swagger.R' 'model-xgboost.R' 'project-param.R' 'project-run.R' 'project-source.R' 'python.R' 'tracking-client.R' 'tracking-experiments.R' 'tracking-observer.R' 'tracking-globals.R' 'tracking-rest.R' 'tracking-runs.R' 'tracking-server.R' 'tracking-ui.R' 'tracking-utils.R'
NeedsCompilation:	no
Packaged:	2025-06-13 05:40:38 UTC; root
Author:	Matei Zaharia [aut, cre], Javier Luraschi [aut], Kevin Kuo [aut], RStudio [cph]
Repository:	CRAN
Date/Publication:	2025-06-18 03:10:02 UTC

mlflow: Interface to 'MLflow'

Description

R interface to 'MLflow', open source platform for the complete machine learning life cycle, see <https://mlflow.org/>. This package supports installing 'MLflow', tracking experiments, creating and running projects, and saving and serving models.

Author(s)

Maintainer: Matei Zaharia matei@databricks.com

Authors:

• Javier Luraschi jluraschi@gmail.com

• Kevin Kuo kevin.kuo@rstudio.com (ORCID)

Other contributors:

• RStudio [copyright holder]

See Also

Useful links:

• https://github.com/mlflow/mlflow

• Report bugs at https://github.com/mlflow/mlflow/issues

Get information from a Databricks job execution context

Description

Parses the data from a job execution context when running on Databricks in a non-interactive mode. This function extracts relevant data that MLflow needs in order to properly utilize the MLflow APIs from this context.

Usage

build_context_tags_from_databricks_job_info(job_info)

Arguments

Value

A list of tags to be set by the run context when creating MLflow runs in the current Databricks Job environment

Get information from Databricks Notebook environment

Description

Retrieves the notebook id, path, url, name, version, and type from the Databricks Notebook execution environment and sets them to a list to be used for setting the configured environment for executing an MLflow run in R from Databricks.

Usage

build_context_tags_from_databricks_notebook_info(notebook_info)

Arguments

Value

A list of tags to be set by the run context when creating MLflow runs in the current Databricks Notebook environment

Initialize an MLflow Client

Description

Initializes and returns an MLflow client that communicates with the tracking server or store at the specified URI.

Usage

mlflow_client(tracking_uri = NULL)

Arguments

Create Experiment

Description

Creates an MLflow experiment and returns its id.

Usage

mlflow_create_experiment(
  name,
  artifact_location = NULL,
  client = NULL,
  tags = NULL
)

Arguments

Create a model version

Description

Create a model version

Usage

mlflow_create_model_version(
  name,
  source,
  run_id = NULL,
  tags = NULL,
  run_link = NULL,
  description = NULL,
  client = NULL
)

Arguments

Create registered model

Description

Creates a new registered model in the model registry

Usage

mlflow_create_registered_model(
  name,
  tags = NULL,
  description = NULL,
  client = NULL
)

Arguments

Delete Experiment

Description

Marks an experiment and associated runs, params, metrics, etc. for deletion. If the experiment uses FileStore, artifacts associated with experiment are also deleted.

Usage

mlflow_delete_experiment(experiment_id, client = NULL)

Arguments

Delete a model version

Description

Delete a model version

Usage

mlflow_delete_model_version(name, version, client = NULL)

Arguments

Delete registered model

Description

Deletes an existing registered model by name

Usage

mlflow_delete_registered_model(name, client = NULL)

Arguments

Delete a Run

Description

Deletes the run with the specified ID.

Usage

mlflow_delete_run(run_id, client = NULL)

Arguments

Delete Tag

Description

Deletes a tag on a run. This is irreversible. Tags are run metadata that can be updated during a run and after a run completes.

Usage

mlflow_delete_tag(key, run_id = NULL, client = NULL)

Arguments

Download Artifacts

Description

Download an artifact file or directory from a run to a local directory if applicable, and return a local path for it.

Usage

mlflow_download_artifacts(path, run_id = NULL, client = NULL)

Arguments

End a Run

Description

Terminates a run. Attempts to end the current active run if 'run_id' is not specified.

Usage

mlflow_end_run(
  status = c("FINISHED", "FAILED", "KILLED"),
  end_time = NULL,
  run_id = NULL,
  client = NULL
)

Arguments

Get Experiment

Description

Gets metadata for an experiment and a list of runs for the experiment. Attempts to obtain the active experiment if both 'experiment_id' and 'name' are unspecified.

Usage

mlflow_get_experiment(experiment_id = NULL, name = NULL, client = NULL)

Arguments

Get latest model versions

Description

Retrieves a list of the latest model versions for a given model.

Usage

mlflow_get_latest_versions(name, stages = list(), client = NULL)

Arguments

Get Metric History

Description

Get a list of all values for the specified metric for a given run.

Usage

mlflow_get_metric_history(metric_key, run_id = NULL, client = NULL)

Arguments

Get a model version

Description

Get a model version

Usage

mlflow_get_model_version(name, version, client = NULL)

Arguments

Get a registered model

Description

Retrieves a registered model from the Model Registry.

Usage

mlflow_get_registered_model(name, client = NULL)

Arguments

Get Run

Description

Gets metadata, params, tags, and metrics for a run. Returns a single value for each metric key: the most recently logged metric value at the largest step.

Usage

mlflow_get_run(run_id = NULL, client = NULL)

Arguments

Get Remote Tracking URI

Description

Gets the remote tracking URI.

Usage

mlflow_get_tracking_uri()

Get Run or Experiment ID

Description

Extracts the ID of the run or experiment.

Usage

mlflow_id(object)

## S3 method for class 'mlflow_run'
mlflow_id(object)

## S3 method for class 'mlflow_experiment'
mlflow_id(object)

Arguments

List Artifacts

Description

Gets a list of artifacts.

Usage

mlflow_list_artifacts(path = NULL, run_id = NULL, client = NULL)

Arguments

Load MLflow Model Flavor

Description

Loads an MLflow model using a specific flavor. This method is called internally by mlflow_load_model, but is exposed for package authors to extend the supported MLflow models. See https://mlflow.org/docs/latest/models.html#storage-format for more info on MLflow model flavors.

Usage

mlflow_load_flavor(flavor, model_path)

Arguments

Load MLflow Model

Description

Loads an MLflow model. MLflow models can have multiple model flavors. Not all flavors / models can be loaded in R. This method by default searches for a flavor supported by R/MLflow.

Usage

mlflow_load_model(model_uri, flavor = NULL, client = mlflow_client())

Arguments

Details

The URI scheme must be supported by MLflow - i.e. there has to be an MLflow artifact repository corresponding to the scheme of the URI. The content is expected to point to a directory containing MLmodel. The following are examples of valid model uris:

- “file:///absolute/path/to/local/model“ - “file:relative/path/to/local/model“ - “s3://my_bucket/path/to/model“ - “runs:/<mlflow_run_id>/run-relative/path/to/model“ - “models:/<model_name>/<model_version>“ - “models:/<model_name>/<stage>“

For more information about supported URI schemes, see the Artifacts Documentation at https://www.mlflow.org/docs/latest/tracking.html#artifact-stores.

Log Artifact

Description

Logs a specific file or directory as an artifact for a run.

Usage

mlflow_log_artifact(path, artifact_path = NULL, run_id = NULL, client = NULL)

Arguments

Details

When logging to Amazon S3, ensure that you have the s3:PutObject, s3:GetObject, s3:ListBucket, and s3:GetBucketLocation permissions on your bucket.

Additionally, at least the AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY environment variables must be set to the corresponding key and secrets provided by Amazon IAM.

Log Batch

Description

Log a batch of metrics, params, and/or tags for a run. The server will respond with an error (non-200 status code) if any data failed to be persisted. In case of error (due to internal server error or an invalid request), partial data may be written.

Usage

mlflow_log_batch(
  metrics = NULL,
  params = NULL,
  tags = NULL,
  run_id = NULL,
  client = NULL
)

Arguments

Log Metric

Description

Logs a metric for a run. Metrics key-value pair that records a single float measure. During a single execution of a run, a particular metric can be logged several times. The MLflow Backend keeps track of historical metric values along two axes: timestamp and step.

Usage

mlflow_log_metric(
  key,
  value,
  timestamp = NULL,
  step = NULL,
  run_id = NULL,
  client = NULL
)

Arguments

Log Model

Description

Logs a model for this run. Similar to 'mlflow_save_model()' but stores model as an artifact within the active run.

Usage

mlflow_log_model(model, artifact_path, ...)

Arguments

Log Parameter

Description

Logs a parameter for a run. Examples are params and hyperparams used for ML training, or constant dates and values used in an ETL pipeline. A param is a STRING key-value pair. For a run, a single parameter is allowed to be logged only once.

Usage

mlflow_log_param(key, value, run_id = NULL, client = NULL)

Arguments

Read Command-Line Parameter

Description

Reads a command-line parameter passed to an MLflow project MLflow allows you to define named, typed input parameters to your R scripts via the mlflow_param API. This is useful for experimentation, e.g. tracking multiple invocations of the same script with different parameters.

Usage

mlflow_param(name, default = NULL, type = NULL, description = NULL)

Arguments

Examples

## Not run: 
# This parametrized script trains a GBM model on the Iris dataset and can be run as an MLflow
# project. You can run this script (assuming it's saved at /some/directory/params_example.R)
# with custom parameters via:
# mlflow_run(entry_point = "params_example.R", uri = "/some/directory",
#   parameters = list(num_trees = 200, learning_rate = 0.1))
install.packages("gbm")
library(mlflow)
library(gbm)
# define and read input parameters
num_trees <- mlflow_param(name = "num_trees", default = 200, type = "integer")
lr <- mlflow_param(name = "learning_rate", default = 0.1, type = "numeric")
# use params to fit a model
ir.adaboost <- gbm(Species ~., data=iris, n.trees=num_trees, shrinkage=lr)

## End(Not run)

Generate Prediction with MLflow Model

Description

Performs prediction over a model loaded using mlflow_load_model(), to be used by package authors to extend the supported MLflow models.

Usage

mlflow_predict(model, data, ...)

Arguments

Register an external MLflow observer

Description

Registers an external MLflow observer that will receive a 'register_tracking_event(event_name, data)' callback on any model tracking event such as "create_run", "delete_run", or "log_metric". Each observer should have a 'register_tracking_event(event_name, data)' callback accepting a character vector 'event_name' specifying the name of the tracking event, and 'data' containing a list of attributes of the event. The callback should be non-blocking, and ideally should complete instantaneously. Any exception thrown from the callback will be ignored.

Usage

mlflow_register_external_observer(observer)

Arguments

Examples

library(mlflow)

observer <- structure(list())
observer$register_tracking_event <- function(event_name, data) {
  print(event_name)
  print(data)
}
mlflow_register_external_observer(observer)

Rename Experiment

Description

Renames an experiment.

Usage

mlflow_rename_experiment(new_name, experiment_id = NULL, client = NULL)

Arguments

Rename a registered model

Description

Renames a model in the Model Registry.

Usage

mlflow_rename_registered_model(name, new_name, client = NULL)

Arguments

Restore Experiment

Description

Restores an experiment marked for deletion. This also restores associated metadata, runs, metrics, and params. If experiment uses FileStore, underlying artifacts associated with experiment are also restored.

Usage

mlflow_restore_experiment(experiment_id, client = NULL)

Arguments

Details

Throws 'RESOURCE_DOES_NOT_EXIST' if the experiment was never created or was permanently deleted.

Restore a Run

Description

Restores the run with the specified ID.

Usage

mlflow_restore_run(run_id, client = NULL)

Arguments

Serve an RFunc MLflow Model

Description

Serves an RFunc MLflow model as a local REST API server. This interface provides similar functionality to “mlflow models serve“ cli command, however, it can only be used to deploy models that include RFunc flavor. The deployed server supports standard mlflow models interface with /ping and /invocation endpoints. In addition, R function models also support deprecated /predict endpoint for generating predictions. The /predict endpoint will be removed in a future version of mlflow.

Usage

mlflow_rfunc_serve(
  model_uri,
  host = "127.0.0.1",
  port = 8090,
  daemonized = FALSE,
  browse = !daemonized,
  ...
)

Arguments

Details

The URI scheme must be supported by MLflow - i.e. there has to be an MLflow artifact repository corresponding to the scheme of the URI. The content is expected to point to a directory containing MLmodel. The following are examples of valid model uris:

- “file:///absolute/path/to/local/model“ - “file:relative/path/to/local/model“ - “s3://my_bucket/path/to/model“ - “runs:/<mlflow_run_id>/run-relative/path/to/model“ - “models:/<model_name>/<model_version>“ - “models:/<model_name>/<stage>“

For more information about supported URI schemes, see the Artifacts Documentation at https://www.mlflow.org/docs/latest/tracking.html#artifact-stores.

Examples

## Not run: 
library(mlflow)

# save simple model with constant prediction
mlflow_save_model(function(df) 1, "mlflow_constant")

# serve an existing model over a web interface
mlflow_rfunc_serve("mlflow_constant")

# request prediction from server
httr::POST("http://127.0.0.1:8090/predict/")

## End(Not run)

Run an MLflow Project

Description

Wrapper for the 'mlflow run' CLI command. See https://www.mlflow.org/docs/latest/cli.html#mlflow-run for more info.

Usage

mlflow_run(
  uri = ".",
  entry_point = NULL,
  version = NULL,
  parameters = NULL,
  experiment_id = NULL,
  experiment_name = NULL,
  backend = NULL,
  backend_config = NULL,
  env_manager = NULL,
  storage_dir = NULL
)

Arguments

Value

The run associated with this run.

Examples

## Not run: 
# This parametrized script trains a GBM model on the Iris dataset and can be run as an MLflow
# project. You can run this script (assuming it's saved at /some/directory/params_example.R)
# with custom parameters via:
# mlflow_run(entry_point = "params_example.R", uri = "/some/directory",
#   parameters = list(num_trees = 200, learning_rate = 0.1))
install.packages("gbm")
library(mlflow)
library(gbm)
# define and read input parameters
num_trees <- mlflow_param(name = "num_trees", default = 200, type = "integer")
lr <- mlflow_param(name = "learning_rate", default = 0.1, type = "numeric")
# use params to fit a model
ir.adaboost <- gbm(Species ~., data=iris, n.trees=num_trees, shrinkage=lr)

## End(Not run)

Save Model for MLflow

Description

Saves model in MLflow format that can later be used for prediction and serving. This method is generic to allow package authors to save custom model types.

Usage

## S3 method for class 'crate'
mlflow_save_model(model, path, model_spec = list(), ...)

mlflow_save_model(model, path, model_spec = list(), ...)

## S3 method for class 'H2OModel'
mlflow_save_model(model, path, model_spec = list(), conda_env = NULL, ...)

## S3 method for class 'keras.engine.training.Model'
mlflow_save_model(model, path, model_spec = list(), conda_env = NULL, ...)

## S3 method for class 'xgb.Booster'
mlflow_save_model(model, path, model_spec = list(), conda_env = NULL, ...)

Arguments

Search Experiments

Description

Search for experiments that satisfy specified criteria.

Usage

mlflow_search_experiments(
  filter = NULL,
  experiment_view_type = c("ACTIVE_ONLY", "DELETED_ONLY", "ALL"),
  max_results = 1000,
  order_by = list(),
  page_token = NULL,
  client = NULL
)

Arguments

List registered models

Description

Retrieves a list of registered models.

Usage

mlflow_search_registered_models(
  filter = NULL,
  max_results = 100,
  order_by = list(),
  page_token = NULL,
  client = NULL
)

Arguments

Search Runs

Description

Search for runs that satisfy expressions. Search expressions can use Metric and Param keys.

Usage

mlflow_search_runs(
  filter = NULL,
  run_view_type = c("ACTIVE_ONLY", "DELETED_ONLY", "ALL"),
  experiment_ids = NULL,
  order_by = list(),
  client = NULL
)

Arguments

Run MLflow Tracking Server

Description

Wrapper for 'mlflow server'.

Usage

mlflow_server(
  file_store = "mlruns",
  default_artifact_root = NULL,
  host = "127.0.0.1",
  port = 5000,
  workers = NULL,
  static_prefix = NULL,
  serve_artifacts = FALSE
)

Arguments

Set Experiment

Description

Sets an experiment as the active experiment. Either the name or ID of the experiment can be provided. If the a name is provided but the experiment does not exist, this function creates an experiment with provided name. Returns the ID of the active experiment.

Usage

mlflow_set_experiment(
  experiment_name = NULL,
  experiment_id = NULL,
  artifact_location = NULL
)

Arguments

Set Experiment Tag

Description

Sets a tag on an experiment with the specified ID. Tags are experiment metadata that can be updated.

Usage

mlflow_set_experiment_tag(key, value, experiment_id = NULL, client = NULL)

Arguments

Set Model version tag

Description

Set a tag for the model version. When stage is set, tag will be set for latest model version of the stage. Setting both version and stage parameter will result in error.

Usage

mlflow_set_model_version_tag(
  name,
  version = NULL,
  key = NULL,
  value = NULL,
  stage = NULL,
  client = NULL
)

Arguments

Set Tag

Description

Sets a tag on a run. Tags are run metadata that can be updated during a run and after a run completes.

Usage

mlflow_set_tag(key, value, run_id = NULL, client = NULL)

Arguments

Set Remote Tracking URI

Description

Specifies the URI to the remote MLflow server that will be used to track experiments.

Usage

mlflow_set_tracking_uri(uri)

Arguments

Source a Script with MLflow Params

Description

This function should not be used interactively. It is designed to be called via 'Rscript' from the terminal or through the MLflow CLI.

Usage

mlflow_source(uri)

Arguments

Start Run

Description

Starts a new run. If 'client' is not provided, this function infers contextual information such as source name and version, and also registers the created run as the active run. If 'client' is provided, no inference is done, and additional arguments such as 'start_time' can be provided.

Usage

mlflow_start_run(
  run_id = NULL,
  experiment_id = NULL,
  start_time = NULL,
  tags = NULL,
  client = NULL,
  nested = FALSE
)

Arguments

Examples

## Not run: 
with(mlflow_start_run(), {
  mlflow_log_metric("test", 10)
})

## End(Not run)

Transition ModelVersion Stage

Description

Transition a model version to a different stage.

Usage

mlflow_transition_model_version_stage(
  name,
  version,
  stage,
  archive_existing_versions = FALSE,
  client = NULL
)

Arguments

Run MLflow User Interface

Description

Launches the MLflow user interface.

Usage

mlflow_ui(client, ...)

Arguments

Examples

## Not run: 
library(mlflow)

# launch mlflow ui locally
mlflow_ui()

# launch mlflow ui for existing mlflow server
mlflow_set_tracking_uri("http://tracking-server:5000")
mlflow_ui()

## End(Not run)

Update model version

Description

Updates a model version

Usage

mlflow_update_model_version(name, version, description, client = NULL)

Arguments

Update a registered model

Description

Updates a model in the Model Registry.

Usage

mlflow_update_registered_model(name, description, client = NULL)

Arguments

Objects exported from other packages

Description

These objects are imported from other packages. Follow the links below to see their documentation.

purrr

%>%

rlang

%||%

zeallot

%<-%